insideHPC » Cloud HPC

How Xoreax Grid Engine Accelerates Applications Through Distributed Processing

Rich Brueckner — Wed, 08 Feb 2012 14:18:39 +0000

www.youtube.com/watch?v=JEtfB5J-mp8

In this video, Xoreax Co-Founder and Chairman Uri Mishol describes how the company’s XGE Grid Engine technology accelerates applications through distributed processing in Windows environments. Recorded at the Israel Supercomputer Conference on Feb. 7, 2012.

Trusted by over 2000 companies and organizations worldwide, XGE combines unique Virtualization technology with robust and flexible distributed computing algorithms to offer a powerful Grid Computing solution for the Windows platform that’s exceptionally simple to deploy and integrate.

Podcast: Opscode Users to Gather for First-Ever #ChefConf Event

Rich Brueckner — Mon, 30 Jan 2012 18:28:59 +0000

In this podcast, Jesse Robbins from OpsCode talks about the new #ChefConf event coming up May 15-17, 2012 in San Francisco. The event will bring together hundreds of users of Chef, an open-source systems integration framework built specifically for automating the cloud.

Tens of thousands of people and thousands of companies use Opscode Chef to automate, manage and scale their infrastructure,” said Jesse Robbins, co-chair of #ChefConf, and co-founder and chief community officer of Opscode. “This ability is now a critical skill for every software developer, systems engineer and IT professional who must manage ever-increasing scale and complexity.”

Download the MP3 * Subscribe on iTunes * If Dropbox is blocked, download from this Google page.

Magellan Study Shows Cloud is 13x More Expensive for Scientific Applications

Rich Brueckner — Thu, 19 Jan 2012 20:01:40 +0000

Does the Cloud really save money? According to the DoE’s Magellan Report relased in December, Public cloud computing environments for scientific applications have the potential to be up to 13 times more expensive than existing DoE supercomputing centers. Read the Full Story.

www.youtube.com/watch?v=oTbYm5f-ByE

In this video first posted here from the September 2011 HPC User Forum, Jeff Broughton from NERSC describes limitations of cloud computing as discovered in the Magellan project.

Fujitsu Launches Cloud Services as a Platform for Big Data

Ralph — Mon, 16 Jan 2012 15:14:31 +0000

Today Fujitsu launched something called Data Utilization Platform Services, which use cloud services as a platform for leveraging big data.

This is a cloud service for gathering, compiling, and integrating massive quantities of sensing data, processing it using realtime processing or batch processing, and using it for such purposes as making future projections. For companies that are considering the use of big data, or quantities of data that are expected to expand to become big data, to develop new products and services or to solve problems, Data Utilization Platform Services offer a way to quickly, safely, and inexpensively get a planned system or business off the ground.

Read the Full Story.

Video: Tricorder X Prize To Integrate with Cloud Computing

Rich Brueckner — Sun, 15 Jan 2012 13:55:37 +0000

www.youtube.com/watch?v=NipbX2wNQ1A

In this video, Peter Diamandis explains the new Qualcomm Tricorder X Prize competition, which aims to reinvent healthcare by spurring creation of integrated diagnostic devices. As envisioned, such a device would integrate today’s exponentially growing technologies including artificial intelligence, wireless sensing, cloud computing, digital imaging, and lab-on-chip technology to enable anyone to diagnose themselves.

Slidecast: IPSoft Overview – Autonomic IT Management

Rich Brueckner — Mon, 09 Jan 2012 12:11:44 +0000

www.youtube.com/watch?v=yDtcPm6rrqs

In this video, Jonathan Crane presents: IPSoft Overview - Autonomic IT Management. Learn more at: http://ipsoft.com

The IT infrastructure of tomorrow will not be managed by people, but by expert systems.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Altair Fires Up HyperWorks On-Demand Cloud

Rich Brueckner — Sat, 17 Dec 2011 17:47:50 +0000

This week Altair Engineering announced it has fired up new datacenter capacity for its HyperWorks On-Demand HPC Cloud. Powered by more than 10,000 cores, the HPC Cloud can support as many as 150 large-scale engineering solver jobs running simultaneously, employing Altair’s solvers RADIOSS, OptiStruct and AcuSolve along with other tools in the HyperWorks family of simulation software.

Companies often turn to HyperWorks On-Demand because they have outgrown their internal capacity or do not have the resources internally to manage high-performance computing equipment,” said Altair Chief Information Officer Martin Nichols. “HyperWorks On-Demand provides all our HyperWorks products as a cloud service, and this data center allows us to scale up to provide much larger on-demand clusters for our customers.”

Altair describes HyperWorks On-Demand as “true HPC” for its end users, meaning that customer simulations are completed as fast as possible. Read the Full Story or check out their new Simulate to Innovate Blog.

Platform Upgrades ISF 3.0 with Cloud Application Designer

Ralph — Wed, 07 Dec 2011 08:33:03 +0000

This week Platform Computing released ISF 3.0 software for building private and hybrid clouds.

Designed to dramatically improve customer productivity when deploying and managing application services, the new version includes the Platform Cloud Application Designer, which allows customers to create reusable templates for deploying business applications in the cloud.

Read the Full Story.

Video: Folding@home: Petaflops Today, Exaflops Soon?

Rich Brueckner — Tue, 06 Dec 2011 20:13:49 +0000

In this video, Vijay Pande from Stanford University presents: Folding@home: Petaflops Today, Exaflops Soon?

This is a fascinating talk if you are at all interested in Alzheimer’s research. After the camera was turned off, I asked Vijay why the brain’s protein folds even though our bodies don’t manufacture new brain cells. He said that we all likely have folded protein in our bodies, but our natural garbage collection defenses normally keep it in check.

Recorded at the HPC Advisory Council Stanford Workshop on Dec. 6, 2011.

Video: BlaZeMeter Startup Offers JMeter Cloud Load Testing at Scale

Rich Brueckner — Tue, 06 Dec 2011 18:55:04 +0000

www.youtube.com/watch?v=BFIcrHkUzGc

In this video, Alon Girmonsky from BlaZeMeter describes the company’s JMeter Cloud for load testing. The new company just completed its Series A funding.

Will Supercomputers Someday Live Only in the Cloud?

Rich Brueckner — Wed, 23 Nov 2011 17:11:08 +0000

David Linthicum write that the new public beta of Cluster Compute Eight Extra Large is Amazon AWS is not only most powerful cloud service yet, it may herald the day when supercomputers live only in the Cloud.

I believe that those who vend these computing systems and sell about 20 to 30 a year will find that the cloud becomes a new and more lucrative channel. Perhaps they will support thousands of users on the cloud, an audience that would typically not be able to afford the hardware and software. Moreover, I believe this might be the only model they support in the future, and the cloud could be the only way to access some platform services. That’s a pity for those who want to see the hardware in their own data center, but perhaps that’s not a bad thing.

I’m really not buying into this prediction, myself. Public Clouds for HPC may work for some, but the speed of light will always keep real supercomputers close to home. Read the Full Story.

Cycle Computing Announces CycleCloud BigScience Challenge Finalists at SC11

Ralph — Thu, 17 Nov 2011 15:02:43 +0000

Cycle Computing has announced the finalists of its CycleCloud BigScience Challenge at SC11. The contest will award $10,000 of computation time to a researcher who is working on behalf of a non-profit organization to further humanity. Finalists were selected based on originality, creativity and suitability to run on CycleCloud clusters launched within Amazon Web Services (AWS). In addition to this prize, Amazon Web Services will provide the winner with a $2,500 credit.

We created the CycleCloud BigScience Challenge to remove boundaries and help democratize access to supercomputing resources” said Jason Stowe, founder and CEO, Cycle Computing. “As a bootstrapped company, we understand why researchers are usually confined to sizing their questions to the compute cluster they have, or can afford. These finalists highlight how utility supercomputing gives scientists the computational room to realize their vision, ask challenging questions, and move humanity forward.”

The finalists:

Alan Aspuru-Guzik, professor in department of chemistry and chemical biology and Johannes Hachmann, postdoctoral fellow, Harvard Clean Energy Project.

Jesus Izaguirre, associate professor of computer science and engineering and concurrent associate professor of applied and computational mathematics and statistics, University of Notre Dame.

Soumya Ray, assistant professor of neurology, Harvard Medical School.

Victor Ruotti, computational biologist, Morgridge Institute for Research.

Martin Steinegger, bioinformatics researcher, TU Munich ROSTLAB.

Each finalist will provide a presentation and demo on their research to the Cycle Judging Panel followed by a 30 minute Q&A. The Finalists’ entries will be judged against the contest criteria and the grand prize winner will be announced next year on the Cycle Computing site.

Visit Cycle Computing for more information on the finalists at Booth #443 at SC11.

Video: The University of Chicago’s Bob Grossman on the Coming Cloud for HPC

Rich Brueckner — Wed, 16 Nov 2011 17:49:49 +0000

www.youtube.com/watch?v=-jPo3tq4D1E

In this video, Bob Grossman from the University of Chicago talks about the state of the Cloud for HPC. Recorded at SC11.

Video: TwinStrata Startup Steps up With Seamless Cloud SANs

RichB — Tue, 15 Nov 2011 14:59:50 +0000

www.youtube.com/watch?v=B7Dj5Hh1HPs

Cloud storage is red hot these days, especially for Startups looking to leverage HPC technologies and innovate their way into the hearts of the enterprise. In this video, Nicos Vekiarides — Chief Executive Officer & Co-Founder of TwinStrata discusses the company’s innovative Cloud San Solutions.

Our core value is Zero-Friction Enterprise Storage, and TwinStrata CloudArray 3.0 removes all friction-points to cloud storage adoption, extending the many advantages of cloud storage to existing storage infrastructures,” said Nicos Vekiarides, Chief Executive Officer at TwinStrata. “CloudArray provides a rapid entry point into cloud storage, without elaborate planning and costly deployments that allows customers to seamlessly roll into public or private clouds in the future.”

Read the Full Story.

Convey and Nimbix Take Hybrid-Cores to the Cloud

Ralph — Tue, 15 Nov 2011 14:34:21 +0000

Convey Computer announced today it is partnering with HPC cloud innovator, Nimbix, to deliver Convey’s unique hybrid-core architecture and bioinformatics personalities as a cloud-based HPC solution. Nimbix, a new Convey customer, will deploy hybrid-core systems to accelerate informatics and discovery applications. Convey has multiple solutions for big data—most notably in bioinformatics and genome research. Convey personalities accelerate applications such as genome sequencing and alignment, giving researchers the ability to solve problems they previously could not. Because Convey’s FPGA-based systems feature an adaptable architecture, customers are creating new personalities every day to attack specific problems they face.

Unlike other cloud infrastructure providers, Nimbix manages accelerated server platforms to speed up data processing for computationally intensive applications, such as bioinformatics, financial analytics, and graphics. In 2010, Nimbix announced its Nimbix Accelerated Compute Cloud, which puts efficient hardware technology, optimized for industry-specific applications, in the hands of customers wanting to solve big data challenges. Nimbix also is expanding its private cloud HPC clusters for customers who prefer dedicated environments for

With our cloud solutions, customers can leverage multiple accelerator technologies, including Convey’s hybrid-core servers, for big data applications. We lower the barrier of access to the ‘new HPC’ via the cloud and deliver a more efficient and productive path to results,” said Steve Hebert, co-founder of Nimbix.

Convey Computer is at Booth #1313 at SC11. Read the Full Story

Video: Dell Fighting Pediatric Cancer with Cloud Resources

Rich Brueckner — Sat, 12 Nov 2011 05:34:43 +0000

This week Dell announced a major commitment of funding, employee engagement and cloud computing technology to support pediatric cancer research programs globally.

As part of its Powering the Possible giving program, Dell is providing the secure cloud-based IT infrastructure to support the world’s first personalized medicine clinical trial for pediatric cancer conducted by the Neuroblastoma and Medullobastoma Translational Research Consortium and supported by The Translational Genomics Research Institute. The cloud infrastructure is both a high-performance computing resource and data exchange/collaboration resource.

I watch the Dell event live on Thursday and it was heartening to see HPC and Cloud Computing making a difference in kid’s lives. Read the Fact Sheet on the Cloud Computing program.

Out of Capacity? Try Cloudbursting with Bright Computing

Rich Brueckner — Wed, 09 Nov 2011 12:07:48 +0000

Last week had the opportunity to sit down with the folks from Bright Computing to talk about their cluster management software. In the wake of a recent announcement that IBM will soon acquire Platform Computing, Bright sees a real opportunity to align with some of Platform partners who might be averse to doing business with Big Blue.

Bright is not sitting still either. This week the company announced the Bright Cluster Manager 6.0, which supports a new capability “cloud bursting.”

For organizations not currently using HPC, Bright Computing’s approach makes HPC more accessible by lowering the barrier to entry. For those already using clusters, Bright makes it fast and easy to extend and manage these systems in the cloud, enabling customers to dynamically add capacity or GPU capabilities as needed.

How does cloud bursting work? In partnership with Amazon Web Services, Bright Cluster Manager enables customers to create new clusters in the cloud, or add cloud-based nodes to existing clusters. With Bright Cluster Manager 6.0, onsite clusters are automatically cloud-ready, at no extra cost.

Bright opened my eyes to a different approach to cloud bursting,” said Addison Snell, CEO of Intersect360 Research. “Bright has found a way to implement cloud as an extension of the local cluster, with a single, consistent interface.”

Bright Computing will demonstrate cloud bursting at Booth #2734 at SC11. Read the Full Story.

Podcast: CycleCloud BigScience Challenge – Win 8 Hours on a 30,000 Core Cluster

Rich Brueckner — Thu, 27 Oct 2011 19:20:28 +0000

Clipped from: blog.cyclecomputing.com (share this clip)

In this podcast, Rich Brueckner from insideHPC interviews Cycle Computing CEO Jason Stowe about the company’s Big Science Challenge 2011. The winner of this contest will receive 8 Hours on 30,000 Core Cluster.

Three quick take-aways:

It’s easy to enter. The submission form has four quick questions.
Finalists will be announced at SC11 and the winner later this year. Cycle Computing will be at Booth #443.
Entries are due Nov. 7, 2011.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players

Video: Big Data, Cloud & Archive Storage: Trends and Opportunities

Rich Brueckner — Thu, 27 Oct 2011 12:00:24 +0000

www.youtube.com/watch?v=I7zEfFVgxOk

In this video, Spectra Logic’s Molly Rector presents: Big Data, Cloud & Archive Storage: Trends and Opportunities. Recorded at the 2011 Spectra Press Day in Boulder.

Slidecast: Riverbed’s High Performance Hybrid Clouds

Rich Brueckner — Tue, 25 Oct 2011 13:02:18 +0000

www.youtube.com/watch?v=gdPQI5FcvJo

In this video, Riverbed’s Nic Rouda presents: High Performance Hybrid Clouds. Today the company rolled out it’s “Stingray” Asymmetric Solutions Portfolio. Acquired from Zeus Technologies, the Stingray family Includes ADC, Web Application Firewall and Web Content Optimization.

Read the Full Story.

Download the MP3 * Subscribe on iTunes * Subscribe on other podcast players